INDEX

Explanations

specific names or terms associated with various subjects, including geographical locations, cultural items, and figures

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ulet

-0.16

ãĥ³ãĥĲ

-0.15

Î·ÏĤ

-0.13

à¸Ńà¸ĩ

-0.13

getCell

-0.13

áº¡n

-0.13

çłĶç©¶æīĢ

-0.13

ìĽĲìĿ´

-0.13

Sai

-0.13

 Giov

-0.12

POSITIVE LOGITS

actory

0.15

mani

0.14

owski

0.13

 Emblem

0.13

.Automation

0.13

 apiUrl

0.13

TOTYPE

0.13

ala

0.13

dec

0.13

ovsky

0.13

Activations Density 0.073%

specific names or terms associated with various subjects, including geographical locations, cultural items, and figures

No Comments

No Known Activations